A case study involving mixture-process variable experiments within a split-plot structure

Tae Yeon Cho, Douglas Montgomery, Connie M. Borror

Research output: Contribution to journalArticle

Abstract

When an experiment involves both mixture and process variables, the size of the experiment can increase dramatically and the assumption of complete randomization may be violated due to the cost or time to change some factor levels. In this situation, restrictions on the randomization of experimental runs are necessary, resulting in a split-plot structure. Furthermore, when some process variables are noise variables, it is important to consider the noise variables at the design stage of the process to find the robust parameter setting that makes the response robust to the variability transmitted from the noise factors. However, many mixture-process experiments are analyzed without considering this randomization issue. We provide a real example of a mixture-process experiment with noise variables within a split-plot structure. Our example demonstrates show to minimize the prediction error with noise variables in a situation where the standard analysis results in poor estimation for the prediction due to the restricted randomization. Without a proper analysis, the experiment leads to the wrong model and results in poor prediction. When the noise variables are ignored in the experiments, the model provides large random errors due to the effect of noise variables. We show that dual optimization using a mean model and a variance model can find the robust settings for the noise variables.

Original languageEnglish (US)
Pages (from-to)80-93
Number of pages14
JournalQuality Engineering
Volume24
Issue number1
DOIs
StatePublished - Jan 1 2012

Fingerprint

Experiments
Random errors
Costs

Keywords

  • experimental design
  • FDS plot
  • mixture-process design
  • noise variable
  • optimality design
  • robust parameter design
  • split-plot design

ASJC Scopus subject areas

  • Safety, Risk, Reliability and Quality
  • Industrial and Manufacturing Engineering

Cite this

A case study involving mixture-process variable experiments within a split-plot structure. / Cho, Tae Yeon; Montgomery, Douglas; Borror, Connie M.

In: Quality Engineering, Vol. 24, No. 1, 01.01.2012, p. 80-93.

Research output: Contribution to journalArticle

@article{d205cf99f1514a0baec494d6f8e66ab6,
title = "A case study involving mixture-process variable experiments within a split-plot structure",
abstract = "When an experiment involves both mixture and process variables, the size of the experiment can increase dramatically and the assumption of complete randomization may be violated due to the cost or time to change some factor levels. In this situation, restrictions on the randomization of experimental runs are necessary, resulting in a split-plot structure. Furthermore, when some process variables are noise variables, it is important to consider the noise variables at the design stage of the process to find the robust parameter setting that makes the response robust to the variability transmitted from the noise factors. However, many mixture-process experiments are analyzed without considering this randomization issue. We provide a real example of a mixture-process experiment with noise variables within a split-plot structure. Our example demonstrates show to minimize the prediction error with noise variables in a situation where the standard analysis results in poor estimation for the prediction due to the restricted randomization. Without a proper analysis, the experiment leads to the wrong model and results in poor prediction. When the noise variables are ignored in the experiments, the model provides large random errors due to the effect of noise variables. We show that dual optimization using a mean model and a variance model can find the robust settings for the noise variables.",
keywords = "experimental design, FDS plot, mixture-process design, noise variable, optimality design, robust parameter design, split-plot design",
author = "Cho, {Tae Yeon} and Douglas Montgomery and Borror, {Connie M.}",
year = "2012",
month = "1",
day = "1",
doi = "10.1080/08982112.2012.627534",
language = "English (US)",
volume = "24",
pages = "80--93",
journal = "Quality Engineering",
issn = "0898-2112",
publisher = "Taylor and Francis Ltd.",
number = "1",

}

TY - JOUR

T1 - A case study involving mixture-process variable experiments within a split-plot structure

AU - Cho, Tae Yeon

AU - Montgomery, Douglas

AU - Borror, Connie M.

PY - 2012/1/1

Y1 - 2012/1/1

N2 - When an experiment involves both mixture and process variables, the size of the experiment can increase dramatically and the assumption of complete randomization may be violated due to the cost or time to change some factor levels. In this situation, restrictions on the randomization of experimental runs are necessary, resulting in a split-plot structure. Furthermore, when some process variables are noise variables, it is important to consider the noise variables at the design stage of the process to find the robust parameter setting that makes the response robust to the variability transmitted from the noise factors. However, many mixture-process experiments are analyzed without considering this randomization issue. We provide a real example of a mixture-process experiment with noise variables within a split-plot structure. Our example demonstrates show to minimize the prediction error with noise variables in a situation where the standard analysis results in poor estimation for the prediction due to the restricted randomization. Without a proper analysis, the experiment leads to the wrong model and results in poor prediction. When the noise variables are ignored in the experiments, the model provides large random errors due to the effect of noise variables. We show that dual optimization using a mean model and a variance model can find the robust settings for the noise variables.

AB - When an experiment involves both mixture and process variables, the size of the experiment can increase dramatically and the assumption of complete randomization may be violated due to the cost or time to change some factor levels. In this situation, restrictions on the randomization of experimental runs are necessary, resulting in a split-plot structure. Furthermore, when some process variables are noise variables, it is important to consider the noise variables at the design stage of the process to find the robust parameter setting that makes the response robust to the variability transmitted from the noise factors. However, many mixture-process experiments are analyzed without considering this randomization issue. We provide a real example of a mixture-process experiment with noise variables within a split-plot structure. Our example demonstrates show to minimize the prediction error with noise variables in a situation where the standard analysis results in poor estimation for the prediction due to the restricted randomization. Without a proper analysis, the experiment leads to the wrong model and results in poor prediction. When the noise variables are ignored in the experiments, the model provides large random errors due to the effect of noise variables. We show that dual optimization using a mean model and a variance model can find the robust settings for the noise variables.

KW - experimental design

KW - FDS plot

KW - mixture-process design

KW - noise variable

KW - optimality design

KW - robust parameter design

KW - split-plot design

UR - http://www.scopus.com/inward/record.url?scp=84857282506&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84857282506&partnerID=8YFLogxK

U2 - 10.1080/08982112.2012.627534

DO - 10.1080/08982112.2012.627534

M3 - Article

AN - SCOPUS:84857282506

VL - 24

SP - 80

EP - 93

JO - Quality Engineering

JF - Quality Engineering

SN - 0898-2112

IS - 1

ER -